Spaces:
Running
on
Zero
Running
on
Zero
Update model/model_manager.py
Browse files- model/model_manager.py +22 -18
model/model_manager.py
CHANGED
@@ -18,6 +18,7 @@ class ModelManager:
|
|
18 |
self.model_vg_list = VIDEO_GENERATION_MODELS
|
19 |
self.excluding_model_list = MUSEUM_UNSUPPORTED_MODELS
|
20 |
self.desired_model_list = DESIRED_APPEAR_MODEL
|
|
|
21 |
self.loaded_models = {}
|
22 |
|
23 |
def load_model_pipe(self, model_name):
|
@@ -28,35 +29,38 @@ class ModelManager:
|
|
28 |
pipe = self.loaded_models[model_name]
|
29 |
return pipe
|
30 |
|
31 |
-
|
32 |
-
def NSFW_filter(self, prompt):
|
33 |
model_id = "meta-llama/Meta-Llama-Guard-2-8B"
|
34 |
-
device = "cuda"
|
35 |
dtype = torch.bfloat16
|
36 |
-
tokenizer = AutoTokenizer.from_pretrained(model_id, token=os.environ['HF_GUARD'])
|
37 |
-
|
|
|
|
|
|
|
38 |
chat = [{"role": "user", "content": prompt}]
|
39 |
-
input_ids = tokenizer.apply_chat_template(chat, return_tensors="pt").to(device)
|
40 |
-
|
|
|
41 |
prompt_len = input_ids.shape[-1]
|
42 |
-
result = tokenizer.decode(output[0][prompt_len:], skip_special_tokens=True)
|
43 |
return result
|
44 |
|
45 |
@spaces.GPU(duration=120)
|
46 |
def generate_image_ig(self, prompt, model_name):
|
47 |
-
|
48 |
-
|
49 |
-
|
50 |
-
|
51 |
-
|
52 |
return result
|
53 |
|
54 |
def generate_image_ig_api(self, prompt, model_name):
|
55 |
-
|
56 |
-
|
57 |
-
|
58 |
-
|
59 |
-
|
60 |
return result
|
61 |
|
62 |
def generate_image_ig_museum(self, model_name):
|
|
|
18 |
self.model_vg_list = VIDEO_GENERATION_MODELS
|
19 |
self.excluding_model_list = MUSEUM_UNSUPPORTED_MODELS
|
20 |
self.desired_model_list = DESIRED_APPEAR_MODEL
|
21 |
+
self.load_guard()
|
22 |
self.loaded_models = {}
|
23 |
|
24 |
def load_model_pipe(self, model_name):
|
|
|
29 |
pipe = self.loaded_models[model_name]
|
30 |
return pipe
|
31 |
|
32 |
+
def load_guard(self)
|
|
|
33 |
model_id = "meta-llama/Meta-Llama-Guard-2-8B"
|
34 |
+
device = "cuda" if torch.cuda.is_available() else "cpu"
|
35 |
dtype = torch.bfloat16
|
36 |
+
self.tokenizer = AutoTokenizer.from_pretrained(model_id, token=os.environ['HF_GUARD'])
|
37 |
+
self.guard = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype=dtype, device_map=device, token=os.environ['HF_GUARD'])
|
38 |
+
|
39 |
+
@spaces.GPU(duration=30)
|
40 |
+
def NSFW_filter(self, prompt):
|
41 |
chat = [{"role": "user", "content": prompt}]
|
42 |
+
input_ids = self.tokenizer.apply_chat_template(chat, return_tensors="pt").to(device)
|
43 |
+
self.guard.cuda()
|
44 |
+
output = self.guard.generate(input_ids=input_ids, max_new_tokens=100, pad_token_id=0)
|
45 |
prompt_len = input_ids.shape[-1]
|
46 |
+
result = self.tokenizer.decode(output[0][prompt_len:], skip_special_tokens=True)
|
47 |
return result
|
48 |
|
49 |
@spaces.GPU(duration=120)
|
50 |
def generate_image_ig(self, prompt, model_name):
|
51 |
+
if self.NSFW_filter(prompt) == 'safe':
|
52 |
+
pipe = self.load_model_pipe(model_name)
|
53 |
+
result = pipe(prompt=prompt)
|
54 |
+
else:
|
55 |
+
result = ''
|
56 |
return result
|
57 |
|
58 |
def generate_image_ig_api(self, prompt, model_name):
|
59 |
+
if self.NSFW_filter(prompt) == 'safe':
|
60 |
+
pipe = self.load_model_pipe(model_name)
|
61 |
+
result = pipe(prompt=prompt)
|
62 |
+
else:
|
63 |
+
result = ''
|
64 |
return result
|
65 |
|
66 |
def generate_image_ig_museum(self, model_name):
|